Background/Purpose

Immunoglobulin A (IgA) autoantibodies (Abs) to citrullinated proteins (ACPAs) are present in the preclinical period of RA development, a finding that suggests a mucosal site of Ab generation (Kokkonen 2011). IgA ACPAs have also been found in subjects at-risk for future RA based on a family history of disease (Arlestig 2012; Barra 2013). In addition, we have identified apparent local lung mucosal generation of IgA RA-related Abs in at-risk subjects (Willis, Demoruelle 2013). Additional findings that serum IgA ACPA predominates during the early natural history of RA would further support initial mucosal generation of RA-related autoimmunity.

Methods

Three groups from the Studies of the Etiology of RA (SERA) project were evaluated: 77 AtRisk subjects who were first-degree relatives of patients with RA or subjects identified through health fair screening and all were serum positive for ³1 Ab including CCP2 (IgG, Axis-Shield), CCP3.1 (IgG/IgA, INOVA) and rheumatoid factor IgA/M/G (INOVA) without inflammatory arthritis, 53 subjects with seropositive RA (1987 criteria), and 71 blood donor controls. Each subject was tested for ACPA IgA/M/G using CCP3 antigen ELISAs (research use only; donated by INOVA), and by a technician blinded to group status. Positivity for CCP isotypes was mean levels +2 standard deviations in a separate set of 70 random blood donors. Shared epitope (SE) status was determined in AtRisk and RA subjects using published methods (Kolfenbach 2009).

Results

When comparing isotype proportions, IgA-CCP was positive in a higher number of AtRisk subjects than IgM (46.8% vs. 27.3%, p=0.03)(Table), and the prevalence was higher than IgG, although this was not statistically significant (46.8% vs. 39.0%; p=0.33). In contrast, in RA IgG-CCP was more common than IgA and IgM (86.8% vs. 69.8% and 47.2%, respectively; p’s²0.03). In Controls, IgA-CCP was also positive in a higher number of subjects than IgM or G (12.7% vs. 2.8% and 1.4%, respectively; p’s²0.03). In AtRisk, smoking was associated with positivity for ³1 CCP isotype; this was not seen in RA, however, RA subjects were more likely to be current smokers than AtRisk. In AtRisk and RA there was no association between CCP isotypes the SE.

Conclusion

The high IgA ACPA positivity in AtRisk and high IgG APCA in RA suggest that mucosal processes may be an early feature of RA-related autoimmunity that later transition to an IgG-dominant process. IgA ACPA positivity in controls also suggests that IgA-CCP is present in wider populations and perhaps related to non-disease-specific immune responses. These findings as well as the association of smoking with ACPA isotypes in AtRisk subjects and current smoking with RA status need exploration in larger studies that examine the genetic, environmental and mucosal factors that may be involved in the evolution of ACPA isotypes in transition from preclinical to clinically apparent RA.